Joint semantics and data-driven path representation for knowledge graph reasoning

被引:9
|
作者
Niu, Guanglin [1 ]
Li, Bo [1 ,2 ]
Zhang, Yongfei [1 ,2 ,3 ]
Sheng, Yongpan [4 ]
Shi, Chuan [5 ]
Li, Jingyang [2 ]
Pu, Shiliang [6 ]
机构
[1] Beihang Univ, Sch Comp Sci & Engn, Beijing Key Lab Digital Media, Beijing, Peoples R China
[2] Beihang Univ, State Key Lab Virtual Real Technol & Syst, Beijing, Peoples R China
[3] Pengcheng Lab, Shenzhen, Peoples R China
[4] Chongqing Univ, Sch Big Data & Software Engn, Chongqing, Peoples R China
[5] Beijing Univ Posts & Telecommun, Beijing, Peoples R China
[6] Hikvision Res Inst, Hangzhou, Peoples R China
基金
中国国家自然科学基金;
关键词
Knowledge graph reasoning; Path representation; Horn rules; Entity converting; Joint semantics and data-driven;
D O I
10.1016/j.neucom.2022.02.011
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Reasoning on a large-scale knowledge graph (KG) is of great importance for KG applications like question answering. The path-based reasoning models can leverage much information over paths other than pure triples in the KG but face several challenges. Firstly, all the existing path-based methods are data-driven, lacking explainability, namely how the path representations and the reasoning results are obtained with human-understandable explanations. Besides, some approaches either consider only relational paths or ignore the heterogeneity between entities and relations both in paths, which cannot capture the rich semantics of paths well. To address the above challenges, in this work, we propose a novel joint semantics and data-driven path representation that balances explainability and generalization in the framework of KG embedding. Specifically, we inject horn rules to obtain the condensed paths through a transparent and explainable path composition procedure. The entity converter is designed to transform entities along paths into the representations in the semantic level similar to relations for reducing the heterogeneity between entities and relations. The KGs, both with and without type information, are considered. Our proposed model is evaluated on two classes of tasks: link prediction and path query answering. The experimental results show that our model obtains significant performance gains over several state-ofthe-art baselines. (c) 2022 Elsevier B.V. All rights reserved.
引用
收藏
页码:249 / 261
页数:13
相关论文
共 50 条
  • [1] A Joint Framework for Explainable Recommendation with Knowledge Reasoning and Graph Representation
    Zhang, Luhao
    Fang, Ruiyu
    Yang, Tianchi
    Hu, Maodi
    Li, Tao
    Shi, Chuan
    Wang, Dong
    [J]. DATABASE SYSTEMS FOR ADVANCED APPLICATIONS, DASFAA 2022, PT III, 2022, : 351 - 363
  • [2] Paleontology Knowledge Graph for Data-Driven Discovery
    Yiying Deng
    Sicun Song
    Junxuan Fan
    Mao Luo
    Le Yao
    Shaochun Dong
    Yukun Shi
    Linna Zhang
    Yue Wang
    Haipeng Xu
    Huiqing Xu
    Yingying Zhao
    Zhaohui Pan
    Zhangshuai Hou
    Xiaoming Li
    Boheng Shen
    Xinran Chen
    Shuhan Zhang
    Xuejin Wu
    Lida Xing
    Qingqing Liang
    Enze Wang
    [J]. Journal of Earth Science, 2024, 35 (03) : 1024 - 1034
  • [3] Paleontology Knowledge Graph for Data-Driven Discovery
    Deng, Yiying
    Song, Sicun
    Fan, Junxuan
    Luo, Mao
    Yao, Le
    Dong, Shaochun
    Shi, Yukun
    Zhang, Linna
    Wang, Yue
    Xu, Haipeng
    Xu, Huiqing
    Zhao, Yingying
    Pan, Zhaohui
    Hou, Zhangshuai
    Li, Xiaoming
    Shen, Boheng
    Chen, Xinran
    Zhang, Shuhan
    Wu, Xuejin
    Xing, Lida
    Liang, Qingqing
    Wang, Enze
    [J]. JOURNAL OF EARTH SCIENCE, 2024, 35 (03) : 1024 - 1034
  • [4] Knowledge graph representation and reasoning
    Cambria, Erik
    Ji, Shaoxiong
    Pan, Shirui
    Yu, Philip S.
    [J]. Neurocomputing, 2021, 461 : 494 - 496
  • [5] Knowledge graph representation and reasoning
    Cambria, Erik
    Ji, Shaoxiong
    Pan, Shirui
    Yu, Philip S.
    [J]. NEUROCOMPUTING, 2021, 461 : 494 - 496
  • [6] A Joint Knowledge Graph Reasoning Method
    Yang, Wenqing
    Li, Xiaochao
    Wang, Peng
    Hou, Jun
    Li, Qianmu
    Zhang, Nan
    [J]. 2022 IEEE INTL CONF ON DEPENDABLE, AUTONOMIC AND SECURE COMPUTING, INTL CONF ON PERVASIVE INTELLIGENCE AND COMPUTING, INTL CONF ON CLOUD AND BIG DATA COMPUTING, INTL CONF ON CYBER SCIENCE AND TECHNOLOGY CONGRESS (DASC/PICOM/CBDCOM/CYBERSCITECH), 2022, : 646 - 651
  • [7] Knowledge-Aided Data-Driven Radar Clutter Representation
    Feng, Yi
    Wongkamthong, Chayut
    Soltani, Mohammadreza
    Ng, Yuting
    Gogineni, Sandeep
    Kang, Bosung
    Pezeshki, Ali
    Calderbank, Robert
    Rangaswamy, Muralidhar
    Tarokh, Vahid
    [J]. 2021 IEEE RADAR CONFERENCE (RADARCONF21): RADAR ON THE MOVE, 2021,
  • [8] (Data-driven) knowledge representation in Industry 4.0 scheduling problems
    Rossit, Daniel A.
    Tohme, Fernando
    [J]. INTERNATIONAL JOURNAL OF COMPUTER INTEGRATED MANUFACTURING, 2022, 35 (10-11) : 1172 - 1187
  • [9] The geoscience knowledge system, ontology and knowledge graph for data-driven discovery: Preface
    Ma, Xiaogang
    Ma, Chao
    Lv, Hairong
    Hu, Xiumian
    [J]. GEOSCIENCE FRONTIERS, 2023, 14 (05)
  • [10] Multi-Source Knowledge Reasoning for Data-Driven IoT Security
    Zhang, Shuqin
    Bai, Guangyao
    Li, Hong
    Liu, Peipei
    Zhang, Minzhi
    Li, Shujun
    [J]. SENSORS, 2021, 21 (22)